Pharmacokinetic and immune stimulating properties of proteins and synthetic drugs may be controlled by their conjugation to certain polymers. For example, polyethylene glycol (PEG) can be conjugated to proteins to achieve this effect (Fee and Van Alstine, Chemical Engineering Science, 61:924-934 (2006)). Such conjugation can take place if the relatively non-reactive hydroxyl groups present in PEG molecules are substituted by other, more reactive moieties (Jagur-Grudzinski, Reactive & Functional Polymers, 39:99-138 (1999)). A standard, linear PEG molecule is chemically a diol, which could suggest that the process of PEG derivatization and purification of products should be trivial. However, the polymeric nature of this diol, together with its amphiphilic properties can make these manipulations difficult. In some cases, the typical laboratory process for separation of difficult reaction mixtures, silica gel-based flash column chromatography, can fail for PEG with molecular weight higher than 1000. Neither crystallization nor precipitation appear adequate to achieve separation of PEG-containing materials, even if these methods can be used for efficient removal of other, contaminating substances with low molecular weight. Most reaction mixtures containing modified PEG molecules lack a reliable analytical method to control or to prove their composition. Polymers with functions that influence only minimally the hydrophobic properties of the polymer can be difficult to analyze by chromatography. The same applies for polymers with functions carrying only a minimal charge. This also applies for preparative chromatographic separation of charged polymers as described elsewhere for the separation of mono- and di-carboxyl modified PEG molecules (Drioli et al., Reactive & Functional Polymers, 48:119-128 (2001)).
Confirmation of results of the synthesis based on NMR can be useless, as long as one is not sure about the purity of the product, and this is typically only obtained by chromatographic methods. This unusual conclusion comes from observations that an equimolar mixture of non-derivatized polymer and bis-derivatized polymer will produce an NMR pattern identical to the pure mono-derivatized polymer. Mass spectrometry can be complicated since most PEG exists not in the form of a single component, but is rather a Gaussian population of different polymer lengths, centered on its average molecular weight. Thus, even if all distinct components of the same type should have their mass increased by the same factor, the presence of unreacted and bis-modified material can obscure the picture of the analysis. The literature discusses this problem only sporadically, and often nothing is mentioned about analysis of the product or its purification. Many authors make the impression that the process that they describe is ongoing with quantitative yield, and thus the quality of the product does not need to be analyzed or questioned. This non-scientific approach can be frequently encountered in the chemistry of PEG. There are many examples in the literature presenting synthetic procedures with four to five consecutive steps without a single analysis of the product at any of these steps, without any attempts to purifying the product, and assuming 100 percent purity at the end of the process. It is, therefore, not strange that researchers after closer testing question these products and their purity (Ananda et al., Anal. Biochem., 374, 231-242 (2008)). A commonly accepted escape from the problem of selective modification is to work with a polymer that has one end blocked from the beginning by a stable chemical group, most often a methoxy group (mPEG). In theory, this blockage converts a PEG molecule to a monofunctional compound, and as such, it could be fully converted to the second derivatized form by increasing the amount of derivatizing reagent and/or time for reaction. Unfortunately, many of reactions commonly applied for derivatization of PEG are sluggish and only seldom go to completion. On the other hand, mPEG preparations contain significant percentages of PEG diol component. Moreover, the amount of this contamination increases with the length of mPEG, and this contamination can be hard to avoid. Consequently, derivatization will also result in formation of symmetrical, bis-derivatizated PEG, and its presence in the conjugating mixture results in formation of cross-linked products with unknown pharmacologic properties or a possible loss of protein activity. Therefore, pure, monofunctional polymers are usually preferred for protein modification, but one should be aware that purification of mPEG from its diol PEG contamination is practically impossible.
Nearly all of existing reactions, used today for derivatization of PEG, belong either to the alkylation-based or the acylation-based category. In the first case, the alkoxy anion, generated from PEG, is reacting with incoming electrophilic modifying reagent. Eventually, the activated PEG, subjected with a good leaving group, is itself an object of a nucleophilic attack. To this category belong processes resulting in thiolation, amination, azidation, and introduction of a carboxyl or an aldehyde group. Modified PEG's of this category will have their functional group connected directly to the PEG terminal carbon atom or these groups will be linked via an ether bond, a thioether bond, or a secondary amino group.
The second category, acylation, is based on a nucleophilic reaction of PEG's hydroxyl, (or another group present in a modified PEG—often an amino group), on an incoming acylating reagent. In many cases, this first acylation is followed by a second acylation that actually introduces the modification of interest to the PEG molecule. Functional groups incorporated by this method can be linked to the rest of PEG by an amido, a carbamido, urethane, thiourethane, or a simple ester group. These linking groups and the chemistry behind them belong to the very traditional methods of combining two chemical identities.
Polyethylene glycols (PEG) coupled to phosphoramidites are used for direct coupling of PEG molecules to synthetic nucleic acids. One example is 4,4′-dimethoxy-trityl-polyethyleneglycol-[(2-cyanoethyl)-(N,N-diisopropyl)]-phosphoramidite. In these compounds, the phosphoramidite group is the part of the reactive functionality for linking the compound to a synthetic nucleic acid. It is designed to work in a completely water-free environment: In the presence of water, the phosphoramidite group can decompose instantaneously, making such PEG phosphoramidites inappropriate for conjugation to biological material in water-containing or aqueous solution. In particular, these PEG phosphoramidites can be inappropriate for conjugation to biological substances which are not soluble, stable or sufficiently reactive in non-aqueous media. Furthermore, already mildly acidic biological substances can decompose these PEG phosphoramidites. Finally, these PEG phosphoramidites contain a labile protecting group adjacent to the phosphorous atom which is specially designed to convert the intermediate phosphotriester to a phosphodiester. Phosphodiesters can be readily degraded enzymatically in vivo.